The Neoadjuvant Rectal (NAR) score is a new surrogate endpoint to be used in clinical trials for early determination of treatment response to different preoperative therapies. The aim is to further validate the NAR-score, primarily developed using chemoradiotherapy (CRT) with a delay to surgery 6-8 weeks, and explore its value using other schedules.
The study included all 9978 patients diagnosed with non-metastasized RC in 2007-2015 that had undergone surgery and was registered in the Swedish Colorectal Cancer Registry. The patients of interest had either short-course radiotherapy (scRT)/CRT+delayed surgery, long-course radiotherapy (RT)+delayed surgery, (C)RT+additional chemotherapy, primary surgery, or scRT+immediate surgery. The scRT/CRT+delayed surgery groups were further divided based on time to surgery.
Mean NAR-score differed significantly (p<0.0001) between different treatments. (C)RT+additional chemotherapy had the lowest mean score of 16.3 and CRT+delayed surgery had 17.7. There was a significant difference (p<0.05) in overall survival (OS) and time to recurrence (TTR) of patients with a Low NAR-score (16) for both CRT- and scRT, with a stronger correlation for CRT-patients. C-index for the NAR-score model (0.623) was not superior to when only pathological T- and N-stage was used (0.646).
The NAR-score is prognostic, but it is not better than pT- and pN-stage. However, the NAR-score can still discriminate between two treatments that have different cell killing effect and may still be of value in clinical trials as an easier method than pT- and N-stage.

Copyright © 2021. Published by Elsevier B.V.

Author